Skip to main content
Thomas M. Selman

Executive Vice President Regulatory Policy

Remarks From the LIMRA/LOMA 2015 Regulatory Compliance Exchange

March 18, 2015

Arlington, VA

On March 30, 1980, the rock and roll band Van Halen held a concert in the gymnasium of the University of Southern Colorado as part of its "Party 'til You Die Tour." Following the concert, the university hosted a dinner for Van Halen, with linens and silverware. According to university officials, the band "proceeded to act like a bunch of animals. They ate the lasagna with their hands, threw the food around the room, smashed the food on the walls and each other." The carpet, drapes and paint in the dining room had to be refurbished. The band's dressing room also was damaged. The university subsequently banned most campus concerts.

Later reports revealed that Van Halen trashed the dining and dressing rooms because of brown M&Ms. Van Halen's contract demanded a variety of munchies, including M&Ms, with the following proviso: "WARNING: ABSOLUTELY NO BROWN ONES." The university had overlooked this clause and served brown M&Ms. For this indignity the band destroyed university property.i

Van Halen thus earned a reputation as insufferable rock and roll musicians, willing to commit gross destruction for a petty grievance. Of course, as a rock and roll band Van Halen reveled in its infamy; the brown M&M caper contributed to its rough image.

Many years later, David Lee Roth, the lead singer, offered an explanation. Roth said that the size and weight of the band's equipment was without precedence. To protect the safety of the band and the audience, the band's contract enumerated the technical requirements for erecting the stage. The contract was a book of specifications. Some promoters might negligently skim it. Roth said that in order to ensure that the promoter read the contract he inserted the brown M&M clause. When the band arrived at each venue he would first check the M&M jar. If it contained brown M&Ms, then he knew that the promoter had not read the contract and that the equipment could be unsafe. The band would send the promoter a message by trashing its dressing room.ii

In other words, Van Halen had inserted a provision to test the promoter's fealty to its contract. Roth had adopted a clever tactic to measure the promoter's behavior. A behavioral economist might say that the promoters who served brown M&Ms displayed overconfidence bias because they were sure that they understood all of the requirements of the lengthy contract without reading it.

Today I'd like to talk about behavioral economics and how its concepts might help regulators and compliance officers. Behavioral economics relies on the cognitive sciences, psychology and neurology, to understand the emotional and intellectual biases that can cause people to make irrational decisions. Behavioral economists also suggest ways in which these biases can be detected, mitigated or redirected toward more rational decisions.

A regulator's application of behavioral economics might help customers make better investment decisions. An understanding of behavioral economics might help ensure that your employees act in an honest and ethical manner. Let me give you some examples of how we can apply economics to help customers, and how you can use the cognitive sciences to encourage good behavior by your employees.

Behavioral economists may suggest two basic methods, "empowerment" and "nudging," to improve an investor's decisions.iii Empowerment describes a set of methods to enhance the ability of people to overcome their biases and make rational, conscious decisions. For example, disclosure might be simplified to "empower" people to make better investment decisions.

A wag once said that "the only people who read prospectuses are those who are paid to do so: lawyers, accountants, and financial advisers." Many investors ignore the dense and turgid prose of the securities prospectus. If disclosure is simplified so that only key items of information are presented, then the average investor might be empowered to read the prospectus, ask questions and make better investment decisions.

This idea is well understood by securities regulators who have attempted to simplify disclosure over the years. For example, in 2009 the SEC simplified the mutual fund prospectus to require key information in plain English in a standardized order. The summary section of the mutual fund prospectus presents information about the fund's investment objectives and strategies, risks, costs, performance and other matters so that the average investor can grasp the essential concepts. As many of you know, there have been similar attempts to simplify the variable annuity prospectus, which could help variable annuity investors better understand the key features of the product.

A cooling off period is another empowerment technique. It helps a person overcome biases that may lead to rash decisions. If an investor must wait before executing on an investment idea, then she is more likely to reflect on her action and make a rational decision. She will not be so easily swayed by strong persuasion. An example of a cooling off period is the time between the filing of a company's registration statement with the SEC and the date it becomes effective. During this cooling off period investors may receive the offer to buy the securities, but must await the effective date before purchasing them.

Empowerment helps the investor overcome natural bias. But sometimes we don't want just to empower investors. We want to nudge them instead, and use their biases to help them make more rational decisions. One example is the automatic enrollment retirement program that is available among employers today. Many investors exhibit inertia; they don't bother to enroll in an employer-sponsored retirement plan. The automatic enrollment programs many companies adopt ensure that employees are enrolled unless they opt out. The automatic enrollment program relies on human inertia to paternalistically "nudge" employees into the employer sponsored plan.

This mechanism is different from the "free look" period in a typical variable annuity. As you know, during the free look period the investor may terminate the contract and retrieve the purchase payments without paying any surrender charges. The automatic enrollment program for retirement programs requires an employee to opt out. If the employee does nothing, then he invests in the plan. The free look period requires an investor to opt in. The investor must act by canceling the VA purchase. Given the inertia tendency, we can assume that most investors allow the free look period to expire. Applying the principles of behavioral economics, we can question whether the free look period leads to rational decision making and whether another construction would lead to a better outcome.

An understanding of these concepts might help regulators and firms to empower investors to make better investment decisions, or nudge them paternalistically into rational choices. These concepts may complement traditional regulation, which some call "command and control." Traditional command and control regulation prescribes behavior to attain a socially desirable objective. "Command and control" regulation is still necessary, but an appreciation of cognitive sciences can help us simplify disclosure and provide other behavioral tools to help investors.

What about our supervision of financial service employees? How might the principles of behavioral economics help us there? Let's begin our inquiry by asking a fundamental question: What do the cognitive sciences tell us about the honesty of a human being? This question has been the subject of many studies by behavioral economists.

Researchers from the University of Oxford and the University of Bonn telephoned over 700 randomly-selected people in Germany, of whom 658 agreed to participate in an experiment. They were told, over the phone, to flip a coin, and to report how it landed. Each person was told that if the coin landed tail's up, he would receive 15 euros or a gift voucher, but if it landed head's up, he would receive nothing. About 56 percent of participants reported that the coin landed head's up, which meant that they would not receive a gift. Only 44 percent reported that the coin landed tails up, in which case they collected the gift. More people, when asked over the phone to report on the coin flip, reported a result that apparently was not in their self-interest.iv

Permit me to cite another study. In 2013 the company Honest Tea set up kiosks throughout the country, offering tea for $1. The kiosks were unmanned. Those who wished to purchase tea had to place their dollar in a box. In every state, at least 80 percent of the subjects paid for their tea. In Hawaii and Alaska, everybody did. In Indiana and Maine, 99 percent did. You might be curious to know which state is the least honest according to this study. It is not a state. It is Washington, DC, apparently the least honest jurisdiction in the country. Should you bemoan our sad state of affairs, be comforted to know that even in Washington, 80 percent paid for their tea.v

A more recent study reported in the Wall Street Journal should give us pause. This study suggests that people may be "situationally honest." A person who is normally honest might be less trustworthy in some situations. This study found that the employees of a large, international bank behaved, on average, honestly in a control condition. However, when reminded of their professional identity as bank employees, it appears that a significant number exhibited signs of dishonesty. The researchers recruited 128 employees, representing many parts of the bank including private banking, asset management, trading, human resources and other support services. They were randomly assigned to either a control group or a treatment group. Those in the control group were asked seven questions that were unrelated to their profession, such as a question about how many hours of television they watch each week. Those in the treatment group were asked seven questions about their professional background, such as the name of the bank and their function at the bank.

All subjects were asked to toss a coin ten times and report the outcomes online. For each coin toss they could win about $20 depending upon whether they reported "heads" or "tails." Of course, we would expect a 50 percent success rate on average, given the probability of tossing either a heads or a tails. The control group reported a 51.6 percent success rate. The treatment group, which had been primed to think about its professional identity as bank employees, reported a 58.2 percent success rate, which is significantly above 50 percent. The researchers repeated the experiment with 133 employees in a broad range of industries. The researchers found that the professional identity questions did not significantly influence the non-bank employees to act more

This study suggests that employees in financial services companies may be tempted to act less honestly at work than they might otherwise. I have only described one study that has reached this conclusion and garnered some attention in the press. I am not a behavioral economist and this study, like the others I described, might be subject to challenge. Nevertheless, firms should be sensitive to those environmental factors that could lead to dishonest behavior. Mitgating or controlling some compensation incentives could facilitate ethical behavior. In 2013, FINRA issued a report on conflicts of interest in the industry and described measures that firms have taken to address those conflicts. The report encouraged broker-dealers to review their compensation practices to eliminate or reduce those incentives that present conflicts of interest.vii

What tools from the cognitive sciences might encourage honest behavior among registered representatives and other employees of your firms?

Let me give you two examples of devices that encourage honest behavior. Both have been proposed by Dr. Dan Ariely, a prominent behavioral economist at Duke University. Ariely has studied the characteristics of dishonesty and how we can encourage trustworthy behavior. One of his studies examined the uses of certification. We assume that when a person signs a representation of certain facts, that the act of signing will help ensure that the representations are accurate. A person's signature can encourage truthfulness like an oath in the courtroom encourages truthful testimony.

Ariely and his colleagues found that the placement of the signature matters. As a person reads a certification form he may quietly decide to be untruthful. But most people must rationalize their dishonesty so that they can feel good about themselves. By the time a person signs the signature at the bottom of the certification, he may have already rationalized his dishonesty. At that point the signature will have no effect. Ariely found that when the signature is required at the beginning of the document then it makes ethics salient when they are most needed, before the signatory has read the document. Signing first is similar to an oath that precedes a witness'

I am sure that there are a variety of certifications that employees of your firms must sign. Your employees might be required to annually attest that they have read, understand and comply with your code of conduct. Billing expense reports are common at any firm. According to Dr. Ariely's research, the signature for these certifications should be placed at the top of the page rather than the bottom, to better ensure that employees honestly report and certify.

This example might not seem significant, but minor alterations to the ways in which compliance tools are applied, in a manner that reflects our understanding of cognitive behavior, might encourage more ethical conduct by employees.

In another study Ariely asked participants either to recall 10 books that they read in high school or the Ten Commandments. The subjects were then enticed to cheat. Those who tried to recall the Ten Commandments did not cheat at all, regardless of their religious commitment. A similar result occurred when students signed a statement that their participation in the study falls under the MIT Honor Code. They did not cheat, even though MIT had no honor code. Ariely's conclusion is that a reminder of a moral code can dissuade a person from engaging in low levels of cheating. The reminder of a moral code, even one to which a person does not adhere, will instill a sense of guilt that encourages honesty.vii

This research suggests that if an employee is reminded of the standards to which he is held, then he is more likely to engage in ethical behavior. As you may be aware, FINRA supports the application of a fiduciary duty to the broker-dealer industry, one that requires broker-dealers and their associated persons to act in their customers' best interest, and that reflects the characteristics of the broker-dealer model. Some in the industry have questioned whether the fiduciary standard would serve much purpose, since many firms already expect their employees to act in their customers' best interest.

Ariely's experiment with the Ten Commandments suggests that the fiduciary duty might offer a subtle but important benefit. If a financial adviser is called "fiduciary" then the mere designation may enforce the behavior that we expect. Like those who tried to recall the Ten Commandments, a financial adviser's acknowledgement that he is a fiduciary with duties to his clients may instill in him the longing to engage in ethical behavior.

I have suggested that our appreciation of the cognitive sciences, of the biases and incentives that investors and financial service employees might have, of the tools that can help us overcome or redirect irrational or dishonest behavior, could supplement normative forms of regulation and compliance. Simplified disclosure, cooling-off periods, minor changes to certifications, and a fiduciary standard are some of the useful tools from behavioral economics.

But we must not enslave ourselves to these concepts. Recall the brown M&Ms. David Lee Roth has said that the prohibition on brown M&Ms was a clever test of whether the promoter had read its contract with Van Halen and carefully erected the concert equipment. There are at least two problems with this story.

First, we can assume that the circle of promoters for Van Halen was small enough that the brown M&Ms trick would be exposed. Every promoter would follow a simple rule: Don't put brown M&M's in Van Halen's dressing room. No rational promoter after the Pueblo, Colorado, concert would do so. In order for this ploy to be most effective, Van Halen should have relied upon a variety of clauses: No brown M&Ms, provide only chocolate milk, always furnish 8 pink roses and 4 white ones. Something like these random, nonsensical clauses might have better served the desired purpose. But a single clause that would quickly become revealed must have been less effective.

There is a second problem with Roth's story, one that could have ghoulish ramifications. Would any of us depend on a brown M&M? If you were David Lee Roth, would you check the M&M jar and having concluded that no brown M&Ms were present assume that the concert stage will not collapse or the wiring will not electrocute the band? Of course not! You would engage in the same meticulous review of the stage set as you would if there were brown M&Ms in the jar. At most, the M&M test can reveal whether the promoter did not read the contract; it cannot tell you that the promoter did.

I have suggested that the cognitive sciences can complement traditional forms of regulation and compliance. The tools of behavioral science can empower or nudge investors toward rational decisions and can encourage honesty in financial service employees. But we cannot depend upon the brown M&M. Rigorous supervision, the expectation of strict adherence to the law and to ethical rules, and enforcement of the rules that we have adopted will remain the hallmark of effective regulation, even as it is enhanced by the provocative conclusions of psychologists, neurologists, and behavioral economists.

Thank you.


i. See
ii. See interview with David Lee Roth at
iii. See generally, Di Porto, F. and N. Rangone, "Behavioural Sciences in Practice: Lessons for EU Policymakers," forthcoming in Nudge and the Law: What Can EU Law Learn from Behavioural Sciences? Hart Publishing, Oxford (September 2015) (manuscript downloaded from
iv. Abeler, J. et al, "Truth-Telling: A Representative Assessment," IZA Discussion Paper No. 6919, Institute for the Study of Labor (October 2012) (manuscript downloaded from
v. (2013).
vi. Cohn, A. et al, "Business culture and dishonesty in the banking industry." Nature 516, pp. 86-89 (December 2014) (manuscript downloaded from
vii. Report on Conflicts of Interest, FINRA (October 2013)
viii. Ariely et al., "Signing at the beginning makes ethics salient and decreases dishonest self-reports in comparison to signing at the end," PNAS pub. (2012) (manuscript downloaded from
ix. Ariely, "Our buggy moral code," (March 2009)